aboutsummaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* new changes per messagetv3 days4-4/+30
|
* asked again countertv3 days3-2/+14
|
* neigbours by transporttv3 days3-1/+36
|
* own tree shapetv3 days3-10/+187
|
* drop full mesh stattv3 days3-11/+7
|
* bench mycelium round tripstv3 days1-0/+32
|
* delta from held headstv3 days4-28/+85
|
* offer again ungeneratedtv3 days2-32/+139
|
* stale delta offerstv3 days2-0/+79
|
* blob window growstv3 days1-5/+10
|
* delta known to sendertv3 days2-6/+14
|
* delta propagationtv3 days2-2/+18
|
* benchmark bridge fast linktv3 days1-17/+22
|
* sim links push backtv3 days1-18/+33
|
* benchmark bridge slow linktv3 days1-8/+19
|
* benchmark bridgetv3 days1-0/+120
|
* sim link ratestv3 days1-1/+49
|
* one delta at a timetv3 days2-9/+53
|
* time blob messagestv3 days8-6/+55
|
* log why a deltatv3 days1-0/+6
|
* offered deltas lapsetv3 days2-7/+52
|
* fix: level means the peer has our headstv3 days1-6/+13
|
* fix: test links fail over cuts by nametv3 days1-5/+6
|
* dashboard: hop delaystv3 days1-3/+35
|
* throughput from delivered messagetv3 days2-2/+59
|
* hop delaystv3 days8-3/+178
|
* test links fail overtv3 days1-0/+105
|
* benchmark members joiningtv3 days1-5/+48
|
* benchmark delays by pathtv3 days3-0/+28
|
* tests count the changes senttv3 days1-5/+7
|
* fix: change publisher stops with the runtimetv3 days1-3/+8
|
* no reset for heads still on their waytv3 days1-1/+44
|
* benchmark what sync bytes aretv3 days4-3/+74
|
* changes from a peer never go backtv3 days2-0/+52
|
* older delivered messages leave us leveltv3 days2-36/+76
|
* benchmark where twice comes fromtv3 days3-6/+20
|
* resend only what is neededtv3 days2-27/+62
|
* benchmark delivered and twicetv3 days3-0/+58
|
* delivered means has all up to our headstv3 days1-13/+59
|
* any message taken as deliveredtv3 days5-40/+112
|
* slow round back offtv3 days5-5/+92
|
* tree messages need no answertv3 days5-7/+235
|
* hello again when heard fromtv3 days3-2/+71
| | | | | | While we wait for a peer's HELLO_REPLY, anything else that comes from it makes our HELLO go again at once, at most once a second, instead of when the backoff says.
* bump pulled_changes_go_along_their_writers_treestv3 days1-0/+3
|
* unanswered pull frees the slottv3 days2-1/+23
| | | | | | When the answer to a pull is overdue (the sync reply timeout: 1 s in the simtests, 10 s in production), the pull gives up the slot, and the next waiting neighbour is pulled at once.
* no heal while both move ontv3 days1-3/+49
| | | | | | | The heal clock now also restarts when our heads and the peer's have both changed since the last STATUS that showed them differing. While writes go on, two neighbours's heads differ because of changes still on their way, and that alone no longer leads to a heal.
* bump pulled_changes_go_along_their_writers_treestv3 days1-2/+52
|
* overdue sync asked after from thentv3 days3-49/+92
|
* pull after gracetv3 days2-4/+33
| | | | | | | | Before pulling, a node waits one measured round trip to the announcer (1 s at most) and pulls only what is still missing. If the round trip isn't known yet, it pulls at once, as now. On the clean benchmark links the round trip is a few milliseconds, so those scenarios shouldn't change.
* announcements answered when lackingtv3 days5-12/+36
| | | | | | | A neighbour now answers an announcement only if it lacks some of the announced heads. Heads that merely differ are no longer enough, since they do while writes go on. The check reads the document, so it runs off the message consumer, as the pull check already does.